Computer Engineering Commons^™

Drone-Based Wireless Communications For Disaster Recovery, Mark Rizko, Cameron Burdsall

Computer Science and Engineering Senior Theses

This project aims to establish a drone system that can deploy a wireless mesh network over a disaster area, which would aid in the process of finding survivors by using wireless communications to identify where victims are and allow authorities to send out alerts to people in the area. We also seek to add a device detection feature that would allow the drones to passively look for devices through WiFi and Bluetooth Low Energy, giving disaster responders the ability to actively identify specific devices, locate zones where victims lie, and discern a rough population estimate of that area. Those that …

Enhanced Sensing Methods For Uav-Based Disaster Recovery, Connor Azzarello, Chris Gerbino, Ruchir Mehta

Computer Science and Engineering Senior Theses

Natural and human-caused disasters devastate and displace civilian populations. Over the past century, the rate at which these catastrophes occur has increased dramatically. Climate change and unsustainable human behaviors are large contributors to the occurrence of natural disasters, therefore it is likely this upward trend will continue. The region of the world where a disaster takes place often determines how severe the implications are for the affected civilians. The devastation that occurs from a disaster is much greater in low-resourced regions of the world.

Unmanned aerial vehicles (UAVs) are commonly used to assist first responders during disaster response. The existing …

Create: Creative Resources To Express Art Through Engagement, Katherine Sanchez, Ari Soriano

Computer Science and Engineering Senior Theses

To address the current mental health crisis, we propose a platform for co-creative systems that users can utilize as art/music therapy. Art therapy can be an effective way of self-expression that results in psychological benefits and improved mood. We analyze the effectiveness of 8 different systems, 4 musical and 4 art-based. These systems allow the users to create while the computer interacts and responds to their input. The results are evaluated on a before and after emotional assessment that we issue to the user. We targeted college age students and distributed our platform to Santa Clara University students during a …

Rebalancing Shared Mobility Systems By User Incentive Scheme Via Reinforcement Learning, Matthew Brian Schofield

Theses and Dissertations

Shared mobility systems regularly suffer from an imbalance of vehicle supply within the system, leading to users being unable to receive service. If such imbalance problems are not mitigated some users will not be serviced. There is an increasing interest in the use of reinforcement learning (RL) techniques for improving the resource supply balance and service level of systems. The goal of these techniques is to produce an effective user incentivization policy scheme to encourage users of a shared mobility system to slightly alter their travel behavior in exchange for a small monetary incentive. These slight changes in user behavior …

Windows Kernel Hijacking Is Not An Option: Memoryranger Comes To The Rescue Again, Igor Korkin

Journal of Digital Forensics, Security and Law

The security of a computer system depends on OS kernel protection. It is crucial to reveal and inspect new attacks on kernel data, as these are used by hackers. The purpose of this paper is to continue research into attacks on dynamically allocated data in the Windows OS kernel and demonstrate the capacity of MemoryRanger to prevent these attacks. This paper discusses three new hijacking attacks on kernel data, which are based on bypassing OS security mechanisms. The first two hijacking attacks result in illegal access to files open in exclusive access. The third attack escalates process privileges, without applying …

Alzheimer’S Disease Diagnostic Support Tool, Chelsea Fernandes, Shreya Venkatesh, Aiyushi Kumar

Computer Science and Engineering Senior Theses

Alzheimer’s Disease is the 6th leading cause of death overall and the most common cause of dementia in older people in the US. The prevalence of the disease is projected to increase in the next few decades and disproportionately impact low/middle income populations. Unfortunately, specialized doctors, such as neurologists, may not be present in situations where a diagnosis is necessary, resulting in the possibility of AD being overlooked at its early and most treatable stages. Our proposed application is a tool that can aid doctors in determining a probable AD diagnosis using an inputted combination of imaging data, biomarkers, patient …

Improved Hyperparameter Tuning For Graph Learning With Warm-Start Configuration, Drew Ligman

Computer Science and Engineering Senior Theses

With the increasing size and complexity of machine learning datasets, obtaining highly performing prediction models in various tasks has become increasingly difficult. In particular, the processs of hyperparameter optimization (HPO) contributes a significant portion of this cost. This work examines a specific graph-machine learning model, graph convolutional networks (GCN), to derive a hyperparameter configuration with optimal performance across a variety of datasets. We motivate our configuration theoretically and validate it empirically through comprehensive experimentation. We find that for GCN semi-supervised classification tasks, our configuration performs nearly optimally when compared against traditional HPO while only requiring a fraction of the budget. …

Bluetooth Security Of Colocated Apps On Android, Sean Kelker, Omar Garcia

Computer Science and Engineering Senior Theses

The Bluetooth protocol is used millions of times per day, as a means of short-range wireless communication. Many of these connections are between a phone running the Android operating system and an external device. Due to Android’s implementation of Bluetooth, however, unrelated applications that are co-located on the phone have the potential to stealthily send and receive communications from any device that is connected. Our project involves creating malicious applications to investigate the effectiveness of this attack on real-world devices, and we show that the vulnerability above has practical applications. We also create a defense against this attack by modifying …

Machine Learning Based Model For The Detection Of Brain Aneurysms From Mr Angiography, Katherine Becknell, Claire Bushnell, Rachel Fitzsimmons, Emily Sumner

Interdisciplinary Design Senior Theses

A brain aneurysm is a thin or weak spot on a blood vessel wall that expands and fills with blood. Brain aneurysms are very dangerous due to the fact that in most cases, patients do not show any symptoms. Because of this, aneurysms are difficult to diagnose unless it becomes very large or ruptures, resulting in fatal hemorrhage.

Aneurysms can be detected by a number of different brain imaging methods including Magnetic Resonance Imaging (MRI), Magnetic Resonance Angiography (MRA), Computed Tomography Angiography (CTA) and other imaging methods but for the sake of this report we will only be focusing on …

Is Social Diversity Related To Misinformation Resistance? An Empirical Study On Social Communities, I Chang, Orion Sun, Jasper Sang Ahn

Computer Science and Engineering Senior Theses

Misinformation has become a pervasive issue on online social media. Whether it is spread purposefully with ill intent or accidentally through ignorance, misinformation can be dangerous and create confusion in those who are affected by it. The investigation presented in this paper found that no research has been performed that directly examine the correlation between a social community’s misinformation resistance and diversity. This project utilizes CrowdTangle, a tool to gather data from Facebook groups, along with machine learning models to determine if this correlation can be drawn. We were able to find correlations between some diversity metrics with misinformation resistance …

Discussing Digital Twin From Of Modeling And Simulation, Zhang Lin, Lu Han

Journal of System Simulation

Abstract: The development and evolution of modeling and simulation technology, and its importance in scientific and technological progress are briefly reviewed. The intrinsic relation between digital twin and modeling and simulation is revealed by analyzing the background and concept of digital twin. The way to build and evaluate a digital twin based on modeling and simulation theoretical methods is discussed. to ensure the credibility.

Development And Prospect Of Simulation Research In China, Xiaogang Qiu, Duan Hong, Xie Xu, Mengna Zhu

Journal of System Simulation

Abstract: It is important to treat modeling and simulation (M&S) as a discipline to advance its development. In China, M&S has been researched and applied over 40 years, and is gradually becoming an independent discipline. Modeling and simulating the complex systems are the challenge, however, this also brings grand opportunity for M&S to widen and improve itself. Since 1980s, the M&S community in China has realized the broad applications of M&S, comprehended M&S from multiple perspectives, conducted extensive research on basic questions, and discussed the composition of the basic simulation theory. By summarizing these achievements, the future works that have …

Multi-View Human Action Recognition Based On Deep Neural Network, Zhao Ying, Lu Yao, Zhang Jian, Qidi Liang, Long Wei

Journal of System Simulation

Abstract: A novel deep neural network named CNN+CA(Convolutional Neural Network plus Context Attention) model is constructed and a new recognition algorithm based on sequence matching is presented to improve the recognition accuracy of MVHAR (Multi-view Human Action Recognition). A CNN(Convolutional Neural Network) is designed to automatically learn multi-view fusion features; the CA (Context Attention) module is introduced to selectively focus on the parts of the features that are relevant for the recognition task; the proposed recognition algorithm based on sequence matching is used to realize MVHAR. The experimental results on the IXMAS dataset and the i3DPost dataset …

Multiple Object Tracking And Kinematic Simulation For Short Track Speed Skating, Li Qi, Hanlin Mo, Xiangdong Wang, Li Hua

Journal of System Simulation

Abstract: Aiming at the problem that it is difficult to obtain the motion data of each athlete in the short track speed skating competition, an algorithm flow of multiple object tracking and kinematic simulation is proposed. A local matching metric is proposed to deal with the partial occlusion in monocular video and improve the tracking stability and robustness. The motion simulation method based on homography mapping and derivative of fitted curve is realized to estimate kinematic parameters such as velocity and acceleration. The experiments on Skating Track Multiple ObjectTracking (STMOT) verified the effectiveness and superiority of the proposed methods.

Spatial Knowledge Representation Model Of Simulation Scenario Based On Ontology, Zhu Jie, Hongjun Zhang

Journal of System Simulation

Abstract: Military simulation scenario has abundant spatial knowledge and is closely related to various models of combat simulation. In view of the lack of uniform specification for the description of simulation scenario knowledge, it is necessary to construct a spatial knowledge representation model in line with spatial thinking to effectively analyze the spatial entities and their interrelationships in simulation scenario. The ontology method is used to establish spatial knowledge domain ontology and form the formal description specification of spatial knowledge concept; the spatial knowledge structure is described hierarchically by using the method of concept knowledge tree, clarifying the semantic logic …

An Artificial Emotion Model For The Mutual Mapping Between Discrete State And Dimensional Space, Zhihang Tian, Xiaming Chen, Dazhi Jiang

Journal of System Simulation

Abstract: Emotional intelligence is an important component and development direction of machine intelligence. The purpose of artificial emotion model is to construct emotion models for machines to develope systematic ability of emotion understanding and expression. However, the existing methods are still insufficient in artificial emotion modeling ability. Aiming at the key factor of personalization in the construction of artificial emotion model, this paper proposes a method of mutual mapping between discrete emotion state and dimension space state, and constructs a machine personalized artificial emotion model based on Big Five personality model and emotion state transfer model. The relevant experimental results …

System Reliability Modeling In Competitive Failure Considering Zoned Shock Effect, Qiguo Hu, Gao Zhan

Journal of System Simulation

Abstract: Shocks have certain effects on system reliability, and how to quantify the degree of shock loads becomes the key of system reliability analysis. The reliability modeling problem of system suffered from different shock strength is researched based on degradation process competing with sudden failure. Shock process is descripted by extreme shock and two types of degradation processes are descripted by Wiener process under the condition of shifting failure threshold. Aiming at different intensity shocks on the system, the different shock load suffered by zoned shock effect is partitioned, the distribution function of the two types of failure processes is …

Decomposition Furnace Outlet Temperature Prediction Based On Elasticnet And Lstm, Guangyu Yu, Xueping Dong, Xiangmin Wang, Gan Min

Journal of System Simulation

Abstract: The outlet temperature of the decomposition furnace is a key indicator in the cement production process. Aiming at the problem that traditional prediction methods only consider the influence of wind, coal, and materials, a temperature prediction model of ElasticNet combined with Long Short-Term Memory (LSTM) neural network is proposed. The ElasticNet-LSTM export temperature prediction model is constructed by using the ElasticNet method to estimate the parameters of different variables, fully considering the influencing factors and realizing the variable screening, and analyzing the influence of the number of hidden layers and nodes on the accuracy of the neural network. Simulation …

A Shared Memory Based Parallel Hierarchical Interest Matching Algorithm, Wenjie Tang, Junwei Cheng, Yiping Yao, Zhu Feng

Journal of System Simulation

Abstract: Interest matching plays an important role in distributed simulation. However, because of huge number of simulation entities and frequent change of regions, interest matching consumes tremendous computation in large scale simulations. The ubiquity of multicore urges us to improve the performance of interest matching by parallelization. A shared memory based parallel hierarchical interest matching algorithm is propose to solve the problem. It maps subscribe regions into a full binary tree, and compares update regions with the tree in parallel. Due to the associative relationship between adjacent nodes, unnecessary comparisons can be eliminated. The experimental results demonstrate that the …

Time-Varying Ocean Channel Modeling Method, Yuehua Pei, Su Wei, Jincheng Tao, Xialin Jiang

Journal of System Simulation

Abstract: For the ocean channel with extremely complicated situation, an underwater acoustic channel modeling algorithm which can reflect the sparsity, time-varying and space-varying characteristics is proposed. Based on the prior information obtained from the sea trials in a specific sea area, the statistical characteristics of the channel impulses response structure with time and space are obtained. According to the obtained channel sparsities and non-zero position vectors’ statistics, a time-varying underwater acoustic channel model match the real environment is generated. Based on the measured data of marine communication in specific sea area, the simulation results show that the proposed …